一维非恒定流模型与卡尔曼滤波耦合的实时交替校正方法研究

Real-time alternative updating model coupling 1-D unsteady flow computation with Kalman filter

本文从一维圣维南方程离散线性化后的差分方程组出发,结合河道水流计算的边界条件,分别以水位、流量及糙率为状态向量建立3种卡尔曼滤波状态空间方程,研发出河道水流计算与变量交替实时校正的耦合模型。与天然河道实测资料类型及系列长短各异的特点相匹配,模型中变量校正模式选择灵活,对每一种校正模式,依据边界条件的给定方式自动切换到相应的处理模块,程序为每一校正时刻、每一校正模块动态地创建相应维数的卡尔曼滤波器。基于环状河网的求解机制,实时交替校正模式被嵌入扩展到环状河网水动力学模型中,初步实现了环状河网水流计算与实时校正的联解。实例检验效果证明了分离变量校正思想的可行性及所建模型的合理性。

Based on the linearized discrete difference equation set obtained from 1-D Saint-Venant equation, a model coupling river flow computation with real-time alternately updating of variance method is developed according to the boundary condition of river flow. In the model three kinds of Kalman filter state spatial equations are established regarding the water level, discharge and roughness as state vectors respectively. In order to conform to the differentiated characteristics of style and number of field data, the selection of variance updating mode in the model is flexible, and every updating model is automatically switched to relevant processing module in terms of setting mode of boundary conditions. The Kalman filter with corresponding dimension is dynamically created for every updating module at every time step. The alternative updating mode is embedded into hydrodynamic model of looping river computation to realize the united solving of river flow computation and real-time updating. The feasibility of this updating idea of discrete variance is verified by an example.